- Email: [email protected]
Age at menarche and blood pressure in pregnancy
Pregnancy Hypertension 15 (2019) 134–140
Contents lists available at ScienceDirect
Pregnancy Hypertension journal homepage: www.elsevier.com/locate/preghy
Age at menarche and blood pressure in pregnancy a,⁎
a,b,c
Clive J. Petry , Ken K. Ong a b c
a
T a
, Ieuan A. Hughes , Carlo L. Acerini , David B. Dunger
a,c
Department of Paediatrics, University of Cambridge, Cambridge, UK Medical Research Council Epidemiology Unit, University of Cambridge, Cambridge, UK The Institute of Metabolic Science, University of Cambridge, Cambridge, UK
A R T I C LE I N FO
A B S T R A C T
Keywords: Gestational hypertension Gestational hypotension Pre-eclampsia Obesity Development
Objectives: To investigate whether age at menarche is related to maternal blood pressure in pregnancy and, if so, whether obesity and insulin resistance can modify the associations. Study design: Analysis of data collected from 438 pregnant women from the longitudinal and prospective Cambridge Baby Growth Study. Main outcome: Testing associations between questionnaire-derived age at menarche and blood pressure measurements in pregnancy collected from hospital notes, and investigating whether any associations were altered by maternal pre-pregnancy body mass index (BMI) and insulin resistance. Measures: Mean arterial blood pressure at four time points across pregnancy, age at menarche, (Homeostasis Model Assessment) insulin resistance around week 28 of pregnancy. Results: For each increased year in age at menarche there was a drop in mean arterial blood pressure (mmHg) of 0.6 at 11.9 weeks, 0.9 at 31.4 and 37.0 weeks, and 0.4 at 38.8 weeks (a maximal difference of over 7 mmHg across extremes of AAM). Each association was attenuated by both maternal pre-pregnancy BMI and insulin resistance. Conclusions: Age at menarche is negatively associated with future blood pressure in pregnancy, so those with the earliest age at menarche have the highest blood pressures. Either these associations may be mediated by links between age at menarche and obesity/insulin resistance, or there may be a confounder (e.g. systemic inflammation) that links age at menarche to each of them.
1. Introduction Age at menarche (AAM) is influenced by a wide range of genetic [1] and non-genetic determinants [2]. Given that some of the genetic variants associated with AAM are also associated with body mass index (BMI) it is perhaps not surprising that AAM is also associated with some BMI-related pathological conditions such as cardiovascular disease [3] and type 2 diabetes [4]. Early AAM, which accompanies most of the increased risk for these conditions, appears to result from increased follicle-stimulating hormone (FSH) and oestradiol concentrations [5], the latter appearing to be continually raised above expected levels into puberty and beyond possibly because of a reduced sensitivity of the hypothalamic-pituitary axis to the negative feedback of circulating steroid hormones. The fact that AAM is also related to the risk of developing another BMI-linked condition in pregnancy, gestational diabetes [6], raises the possibility that the hormonal changes involved in altering AAM may also affect pregnancy hormone concentrations. Indeed, there is in vitro
⁎
evidence from cultured bovine granulosa cells that FSH is able to alter expression of pregnancy-associated plasma protein A (PAPP-A) [7]. We recently found non-linear associations between AAM and serum PAPPA concentrations [8], which could be important because of the potential role of PAPP-A in regulating pregnancy insulin sensitivity [9]. In our study AAM was negatively associated with insulin resistance in pregnancy [8]. Given the links between AAM and cardiovascular disease [3], pre-eclampsia and cardiovascular disease [10] and pre-eclampsia with insulin resistance [11], it is not surprising that AAM has been found to be negatively associated with pre-eclampsia in most [12–15] although not all [16] studies. The first sign of pre-eclampsia is often raised blood pressure. Although AAM is associated with various cardiovascular traits outside of pregnancy its association with blood pressure in pregnancy has not been thoroughly investigated. This study was therefore implemented to investigate potential associations between AAM and blood pressure across pregnancy.
Corresponding author at: Department of Paediatrics, Box 116, Cambridge Biomedical Campus, Hills Road, Cambridge CB2 0QQ, UK. E-mail address: [email protected] (C.J. Petry).
https://doi.org/10.1016/j.preghy.2019.01.004 Received 21 September 2018; Received in revised form 11 December 2018; Accepted 10 January 2019 Available online 11 January 2019 2210-7789/ © 2019 International Society for the Study of Hypertension in Pregnancy. Published by Elsevier B.V. All rights reserved.
Pregnancy Hypertension 15 (2019) 134–140
C.J. Petry et al.
2. Materials and methods
guidelines. Written informed consent was obtained from all the study participants.
2.1. Cohort 2.6. Assays Between 2001 and 2009 the prospective and longitudinal Cambridge Baby Growth Study recruited 2229 mothers (and their and offspring) attending ultrasound clinics during early pregnancy at the Rosie Maternity Hospital, Cambridge, United Kingdom [9]. The study participants, all of whom were over 16 years of age and none of whom reported suffering from chronic hypertension or were treated with antihypertensives prior to conception, received routine care during pregnancy including the monitoring of their blood pressure. In addition fasting blood samples were collected from 1239 participants (55.6%) for the measurement of plasma glucose and insulin concentrations around week 28 of pregnancy for the evaluation of insulin resistance by the Homeostasis Model Assessment (HOMA) [17]. In this cohort, 96.9% of the offspring were of white ethnicity, 0.9% were Indo-Asian, 0.8% were East-Asian, 0.6% were black (African or Caribbean) and a further 0.8% were of mixed race.
PAPP-A was measured by time-resolved fluoroimmunoassay (autoDELFIA, Perkin Elmer Ltd., Seer Green, U.K.). The minimum detection limit of this assay was 5 mIU/L. The intra- and inter-assay coefficients of variation (CV) was less than 8% and 10% throughout. Insulin concentrations were measured by enzyme-linked immunosorbent assay using a commercial kit (DSL, London, U.K.). The minimum detection limit of this assay was 1.6 pmol/L. Intra-assay CVs were 4.4 and 5.1% at 62 pmol/L and 215 pmol/L, and equivalent interassay CVs were 8.7 and 2.9%; this assay has no cross-reactivity with proinsulin at levels up to 1,000 pmol/L. Blood glucose concentrations were measured using a routine glucose oxidase-based method. 2.7. Calculations The BMI before pregnancy was calculated as the pre-pregnancy body weight divided by the height squared. The reproductive age was determined as the maternal age at the time of the birth of her baby in the index pregnancy minus the AAM. Mean arterial blood pressure in pregnancy was calculated as the sum of twice the diastolic plus the systolic blood pressure, divided by three. HOMA IR was estimated from the fasting glucose and insulin concentrations using the HOMA calculator available at https://www.dtu.ox.ac.uk/homacalculator/ [17]. The index of multiple deprivation was derived from residential post codes as described [20].
2.2. Age at menarche Each of the study participants was given a printed questionnaire at recruitment to complete during the rest of the pregnancy [8], of which a total of 1273 women (57.1%) did so. One of the questions asked “What age were you when you had your first period?” The mean (95% confidence interval) AAM of those women was 12.9 (12.9, 13.0) years. 2.3. Blood pressure during pregnancy
2.8. Statistical analysis
Records of routine blood pressure measurements made during pregnancy were collected from the hospital notes from 968 women in the Cambridge Baby Growth Study (hospital notes for the other study participants either not being available or not containing record of these blood pressures) [18]. Of all the readings four times in pregnancy emerged as points when blood pressure was routinely measured in the majority of subjects. According to the gestational week at which the measurements were taken, the first three time points were: (1) at 11.8 (11.5, 12.0) weeks, (2) at 31.4 (31.3, 31.5) weeks and (3) at 37.0 (36.9, 37.0) weeks. The fourth readings were taken specifically during the final 2 weeks prior to parturition (mean 38.8 (38.7, 38.9) weeks). The present analysis was restricted to data from 438 pregnancies where all four blood pressure measurements and the AAM were available.
Associations between (mean arterial, systolic or diastolic) blood pressure and AAM were assessed by linear regression (adjusted only for the exact stage of pregnancy or both that and either maternal prepregnancy BMI or week 28 HOMA IR). Associations with serial blood pressure measurements were analysed by general estimation equation modelling, adjusting for weeks of gestation ( ± BMI or HOMA IR) when the blood pressure readings were taken. Associations between other continuous variables (adjusted for confounders where appropriate) were also assessed by linear regression whereas associations with HDP or gestational hypotension were assessed using logistic regression. Associations between categorical variables were tested using the χ2test. Unless otherwise stated all other data are presented as means (95% confidence intervals). Statistical analyses were performed using Stata 13 (StataCorp LP, College Station, Texas, U.S.A.). P < 0.05 was considered statistically significant throughout.
2.4. Gestational hypertension & hypotension When hospital notes were investigated additional evidence of hypertensive disorders of pregnancy (HDP) was also sought (defined using the inclusion of a diagnosis of the following terms: “pre-eclampsia”, “gestational hypertension” or “pregnancy-induced hypertension”). In addition further women were categorised as having HDP using the National Institute for Health and Care Excellence criteria for gestational hypertension (blood pressure measurements in the second half of pregnancy that are > = 140 mmHg systolic or 90 mmHg diastolic blood pressure in women without chronic hypertension) [19], with the exception that for our study evidence of HDP was accepted if the blood pressure cut offs were exceeded at one reading rather than at least two [18]. Evidence of gestational hypotension was defined in this study as any one of the blood pressure recordings for a study participant being < = 110/60 mmHg.
3. Results 3.1. Study participants There were no differences between those from the Cambridge Baby Growth Study that were included in the present analysis and those that were not in terms of maternal age, pre-pregnancy BMI, blood pressures in the second half of pregnancy, week 15 PAPP-A concentrations, week 28 glucose and insulin concentrations and HOMA IR, and the baby’s birth weight adjusted for known confounders (Table 1). However fewer of those included in the present analysis smoked in pregnancy and they had lower blood pressures at booking clinic, higher gestational ages at the birth of their babies (of an average of around 4 days) and higher unadjusted baby birth weights as well as lower indexes of multiple deprivation. A higher proportion were also nulliparous and developed HDP, although there was no difference in the proportion that developed pre-eclampsia. Mean arterial blood pressure at booking clinic was negatively associated with AAM (see below) in the women included in the
2.5. Ethics The Cambridge Baby Growth Study was approved by the local ethics committee, Addenbrooke’s Hospital, Cambridge, United Kingdom. All procedures followed were in accordance with the institutional 135
Pregnancy Hypertension 15 (2019) 134–140
3.2. Associations between AAM and blood pressure in pregnancy At booking clinic AAM showed a negative association with mean arterial blood pressure (Table 2). By weeks 31 and 37 the effect sizes of the negative associations were greater in terms of effect size and significance. Just prior to the birth of their baby the effect size of the 136
0.5 −0.3 (–1.1, +0.6) 308 0.2 −0.5 (−1.4, +0.3) 0.9 0.3
402
−0.1 (–0.6, +0.8)
308
0.2 0.06 −0.4 (–1.1, +0.3) −0.7 (–1.5, 0) 308 308 0.07 9.6 × 10−3 −0.7 (–1.4, 0) −1.1 (–1.8, –0.3) 308 308 0.1 0.1 −0.6 (–1.0, +0.1) −0.5 (–1.1, +0.2) 402 402 2.0 × 10−3 4.2 × 10−3
−0.1 (–0.8, +0.6) 308 0.3 −0.3 (–1.1, +0.4) 308 0.7 −0.1 (–0.7, +0.5) 402 0.07
Change in Mean Arterial Blood Pressure (mmHg/year) N N p-value
402 0.2
Data are means (95% confidence interval). Model 1 is adjusted for gestational age when the blood pressures were measured. Model 2 is adjusted for gestational age when the blood pressures were measured in women with pre-pregnancy BMI recorded. Model 3 is adjusted for gestational age and pre-pregnancy BMI when the blood pressures were measured in women with BMI recorded. Model 4 is adjusted for gestational age when the blood pressures were measured in women with week 28 HOMA IR recorded. Model 5 is adjusted for gestational age and week 28 HOMA IR when the blood pressures were measured in women with HOMA IR recorded.
present analysis, but there was no significant association of AAM with smoking in pregnancy (p = 0.2), gestational age at the birth of their babies (p = 1.0), unadjusted birth weights (p = 0.2), index of multiple deprivation (p = 0.2) or parity (p = 0.08).
438
Data are means (95% confidence intervals). *adjusted for gestational age at birth, sex, parity and maternal pre-pregnancy BMI.
Change in Mean Arterial Blood Pressure (mmHg/year)
p-value
0.4
38.8 (38.7, 38.9)
3.469 (3.438, 3.500) (n = 779)
402 402
Baby’s adjusted birth weight (kg)*
5.1 × 10–5
1.7 × 10−3 7.1 × 10−3
Baby’s unadjusted birth weight (kg)
3.6 × 10–10
438 438
Gestational age of baby at birth (weeks)
0.5
31.4 (31.3, 31.5) 37.0 (36.9, 37.1)
Week 28 HOMA IR (%)
0.3
−0.6 (–1.2, 0) −1.0 (–1.6, –0.4) −0.9 (−1.6, −0.3) −0.4 (−1.1, +0.3)
Week 28 fasting insulin concentration (pmol/L)
0.7
402
Week 28 fasting glucose concentration (mmol/L)
0.3
0.05
Week 15 serum PAPP-A concentration (mIU/L)
0.5
−0.6 (–1.2, 0) −0.9 (–1.3, –0.3) −0.9 (−1.5, −0.2) −0.4 (−1.1, +0.3)
Pre-eclampsia (n)
0.01
438
HDP (n)
0.8
11.9 (11.6, 12.1)
Mean arterial blood pressure (mmHg) at 38.8 weeks
0.6
Change in Mean Arterial Blood Pressure (mmHg/year)
Mean arterial blood pressure (mmHg) at 37.0 weeks
0.3
p-value
Mean arterial blood pressure (mmHg) at 31.4 weeks
0.02
Model 4
Mean arterial blood pressure (mmHg) at 11.7 weeks
6.6 × 10–3
Model 3
Index of multiple deprivation
0.03
Model 2
Smoked during pregnancy (n)
0.01
Model 1
Parity (n nulliparous)
0.3
Table 2 Associations between AAM and mean arterial blood pressure in pregnancy in the Cambridge Baby Growth Study in women for whom all data were available.
Pre-pregnancy BMI (kg/m2)
N
0.5
Change in Mean Arterial Blood Pressure (mmHg/year)
33.6 (33.3, 33.8) (n = 908) 24.2 (23.9, 24.5) (n = 785) 503 out of 1208 (41.6%) 72 out of 1218 (5.9%) 11.1 (9.7, 12.6) (n = 40) 81.7 (80.7, 82.8) (n = 260) 82.0 (81.0, 83.1) (n = 265) 85.2 (84.0, 86.4) (n = 247) 86.6 (84.3, 88.9) (n = 90) 19 out of 530 (3.6%) 12 out of 530 (2.3%) 6879 (6331, 7427) (n = 582) 4.3 (4.3, 4.4) (n = 770) 53 (50, 56) (n = 806) 93 (89, 96) (n = 749) 39.6 (39.5, 39.7) (n = 1,219) 3.448 (3.417, 3.478) (n = 1,213)
N
33.4 (33.0, 33.8) (n = 427) 23.9 (23.4, 24.3) (n = 402) 214 out of 438 (48.9%) 14 out of 437 (3.2%) 9.0 (8.5, 9.5) (n = 321) 80.1 (79.2, 80.9) (n = 438) 81.4 (80.6, 82.2) (n = 438) 85.0 (83.9, 85.7) (n = 438) 87.0 (86.0, 88.0) (n = 438) 32 out of 438 (7.3%) 7 out of 438 (1.6%) 6999 (6143, 7854) (n = 239) 4.3 (4.3, 4.4) (n = 313) 52 (48, 56) (n = 330) 90 (85, 96) (n = 310) 40.2 (40.0, 40.3) (n = 438) 3.563 (3.513, 3.614) (n = 436) 3.494 (3.451, 3.538) (n = 401)
p-value
Maternal age (years)
Change in Mean Arterial Blood Pressure (mmHg/year)
p-value
N
Excluded women
Gestational Age (Weeks)
Included women
Model 5
Characteristic
p-value
Table 1 Characteristics of those Cambridge Baby Growth Study participants who were part of this analysis (from whom we had self-reported AAM and at least 4 blood pressures in pregnancy recorded from hospital notes) and comparison with those participants that were not included in the analysis.
0.7
C.J. Petry et al.
Pregnancy Hypertension 15 (2019) 134–140
C.J. Petry et al.
Table 3 Associations between serial mean arterial blood pressures (4 measurements taken over pregnancy) and AAM as analysed by general estimation equation modelling. Mean Arterial Blood Pressure Effect Size (mmHg/year)
p-value
n
Confounders in model
−0.7 −0.6 −0.7 −0.2 −0.6 −0.4
6.3 × 10−3 7.9 × 10−3 7.1 × 10−3 0.4 0.05 0.2
438 438 402 402 308 308
Gestational Gestational Gestational Gestational Gestational Gestational
(−1.2, (−1.0, (−1.2, (−0.7, (−1.2, (−0.9,
−0.2) −0.2) −0.2) 0.3) 0) 0.2)
age age age age age age
when when when when when when
blood blood blood blood blood blood
pressures pressures pressures pressures pressures pressures
measured, measured, measured, measured, measured, measured,
identity identity identity identity identity identity
number number, HDP, pre-eclampsia number (if BMI data available) number, BMI number (if week 28 HOMA IR data available) number, week 28 HOMA IR
Data are mean (95% confidence interval).
Fig. 1. Predicted systolic blood pressures (after adjustment for gestational age) against AAM in linear regression models from the Cambridge Baby Growth Study: at 11.9 weeks (top left, β = −1.0 (−1.9, −0.2) mmHg/year, p = 0.02), 31.4 weeks (top right, β = −1.1 (–1.9, –0.3) mmHg/year, p = 9.2 × 10−3), 37.0 weeks (bottom left, β = −1.2 (−2.0, −0.4) mmHg/year, p = 5.3 × 10−3) and 38.8 weeks (bottom right, β = −0.8 (−1.6, +0.1) mmHg/year, p = 0.08). N = 438 throughout. Solid lines are the means and the dashed lines are the 95% confidence intervals.
n = 427).
association was smaller (and no longer statistically significant), albeit in the same direction as the earlier associations in pregnancy. Overall using repeat measures there was a significant negative association between AAM and mean arterial blood pressure (Table 3), which was attenuated by both pre-pregnancy BMI and HOMA IR as were the individual associations (Table 2). The relationship between blood pressure and AAM appeared to be stronger with systolic (Fig. 1) rather than diastolic blood pressure (Supplementary Table 1). The reproductive age was not associated with mean arterial blood pressure at any of the four blood pressure readings: 11.9 weeks (β = 0.0 (-0.2, 0.2) mmHg/year, p = 0.9, n = 427), 31.4 weeks (β = -0.1 (-0.3, 0) mmHg/year, p = 0.1, n = 427), 37.0 weeks (β = -0.1 (-0.3, +0.1) mmHg/year, p = 0.3, n = 427), 38.8 weeks (β = -0.2 (-0.4, 0) mmHg/year, p = 0.08,
3.3. Associations between AAM, HDP and gestational hypotension There was no significant association between AAM and HDP (odds ratio (OR) 0.89 (0.67, 1.19) per year, p = 0.5, n = 438), even though they were associated with both HOMA IR (OR 1.6 (1.1, 2.3) per HOMA IR unit, p = 0.03, n = 335) and BMI (OR 1.2 (1.1, 1.3) per kg/m2, p = 5.4 × 10–5, n = 414) in these women. There was also no significant difference in AAM between those that had HDP and those that did not: no HDP 13.0 (12.8, 13.1) years (n = 413) v. HDP 12.8 (12.2, 13.3) years (n = 25) (p = 0.5). There was a significant positive association between AAM and 137
Pregnancy Hypertension 15 (2019) 134–140
C.J. Petry et al.
adjusted for either pre-pregnancy BMI or HOMA IR (both pre-pregnancy BMI and HOMA IR [8] themselves being strongly negatively associated with AAM), suggesting a potential mechanism involving increased adiposity and insulin resistance underpinning the link between AAM and blood pressure in pregnancy. Insulin resistance and obesity are well known associates of increased blood pressure both in and outside of pregnancy [21]. Evolutionary biology suggests that the insulin resistance may not be causal in this relationship however. Rather both it and increased blood pressure may result from the low grade systemic inflammation occurring as a result of the increased macrophage infiltration of adipose tissue [22] and other tissues containing ectopically-stored lipids [23] that occurs in obesity [24]. Consistent with this AAM has been reported to be negatively associated with circulating C-reactive protein concentrations suggesting a link between early AAM and inflammation [25]. Such effects appear to arise from the interaction between historically beneficial “thrifty” genotypes (that favour energy storage) and modern diets and lifestyles. In pregnancy placental hormones can be part of these processes. In the Cambridge Baby Growth Study we previously found that low week 15 PAPP-A concentrations were associated with insulin resistance in pregnancy [9] and that women with an early AAM had lower PAPP-A concentrations than women with AAMs closer to the average [8]. In the present analysis week 15 PAPP-A concentrations were negatively associated with all the blood pressure measurements. Further studies are needed to identify whether PAPP-A is causal in these processes or whether the associations observed are confounded, possibly by the insulin resistance resulting from the low grade systemic inflammation described above. The associations between early AAM and raised blood pressure in pregnancy observed in this study could be part of a particular female life course trajectory framework that is “high risk” for poor cardiometabolic health. AAM has been related to each of the steps in the following sequence: the combination of low birth weight and subsequent catch-up growth in infancy [26], or alternatively in utero exposure to gestational diabetes [27], followed by childhood obesity [28], early AAM and subsequent epiphyseal fusion [29], shorter adult height and adult obesity [8], insulin resistance [8], raised blood pressure (the present study; possibly increasing the risk of pre-eclampsia [12–15]) and increased glucose concentrations [8] in pregnancy, cardiovascular disease [3,30] and type 2 diabetes [4] in later life. Although somewhat speculative there is evidence of links between the other factors [31] even if you remove early AAM from the sequence. However these and the present study suggest that AAM should be considered when deliberating over the developmental origins of health and disease. Although the present study has found apparently robust negative associations between AAM and blood pressure in pregnancy it does have a number of limitations. The first of these is that the women tested may not be entirely representative of the population from which they were sampled. The Cambridge Baby Growth Study cohort as a whole is representative of the South Cambridgeshire population of pregnant women, albeit on average the mothers being slightly older and more likely to be nulliparous [32]. However the subpopulation from whom the present analysis was drawn included women whose blood pressure at booking clinic was slightly lower than that of the cohort as a whole, with their baby’s births occurring on average 4 days later than those not included leading to increased unadjusted birth weights of the babies. A lower proportion of the mothers included in the present analysis smoked (although smoking was uncommon as a whole in this population) and they were less deprived in a population not noted for high levels of deprivation. A higher proportion of them were also nulliparous and developed HDP. However the only one of these that was significantly associated with AAM was the booking clinic blood pressure so any biases may have been limited. Another limitation to the study is that the AAM, at a time when AAMs were thought to be declining in the U.K. at a slow rate [33], was self-reported so some of them may have been inaccurate. However it has previously been reported that in a generally well-qualified population, as is a characteristic of the
Table 4 Associations between circulating PAPP-A concentrations from around week 15 of pregnancy and mean arterial blood pressures in those pregnancies. Gestational Age (Weeks)
Change in Mean Arterial Blood Pressure (SDs/year)
N
p-value
11.9 31.4 37.0 38.8
−0.182 −0.188 −0.206 −0.108
261 261 261 261
0.01 0.01 4.9 × 10−3 0.1
(11.6, (31.3, (36.9, (38.7,
12.1) 31.5) 37.1) 38.9)
Data are means (with 95% confidence intervals for gestational ages).
unadjusted evidence of gestational hypotension (OR 1.2 (1.1, 1.4) per year, p = 2.3 × 10−3, n = 438). Those women who experienced hypotension in pregnancy had significantly later AAMs than those that did not: no hypotension 12.7 (12.5, 12.9) years (n = 193) v. hypotension 13.2 (13.0, 13.3) years (n = 245) (p = 2.0 × 10−3). 3.4. Associations between PAPP-A concentrations, HOMA IR and blood pressure in pregnancy Blood pressure in pregnancy was negatively associated with serum PAPP-A concentrations from around week 15 of pregnancy (Table 4), with the first three measurements having statistically significant associations and the association with the late pregnancy measurement having a relatively smaller (not statistically significant) effect size. All the blood pressure measurements were also positively associated with week 28 HOMA IR (Table 5). We previously reported associations between AAM and: (1) pre-pregnancy BMI, (2) week 15 circulating PAPPA concentrations and (3) week 28 HOMA IR in our population [8]. 4. Discussion In this study we have shown negative associations between AAM and four blood pressure measurements in pregnancy. Whilst only three of the associations reached statistical significance, two were strong and all of them plus the combined association were consistently in the same direction with just the effect size altering. The highest blood pressures were consistently in those women with early AAM. The effect sizes were not large enough to independently change the risk for HDP in our population, although AAM was positively associated with evidence of gestational hypotension. The lack of association with HDP or even preeclampsia (data not shown) could help explain why AAM is not associated with pre-eclampsia in every study [16]. However with those women with the earliest AAMs having on average predicted mean arterial blood pressure up to over 7 mmHg higher than those with the latest AAMs in our population, factors related to AAM could clearly be associated with changes in risk in combination with other risk factors. The same does not appear to be true for the time elapsed between the AAM and the mother’s age at the time of the birth of her baby. Associations between AAM and mean arterial blood pressure were slightly weaker in the reduced number of women for whom we had prepregnancy BMIs available, presumably due to the accompanying reduction in statistical power. They were attenuated when they were Table 5 Associations between HOMA IR (fasting insulin resistance) from around week 28 of pregnancy and mean arterial blood pressures in those pregnancies. Gestational Age (Weeks)
Change in Mean Arterial Blood Pressure (SDs/year)
N
p-value
11.9 31.4 37.0 38.8
0.221 0.301 0.314 0.235
335 335 335 335
4.8 × 10−5 1.7 × 10−8 2.7 × 10−9 1.6 × 10−5
(11.6, (31.3, (36.9, (38.7,
12.1) 31.5) 37.1) 38.9)
Data are means (with 95% confidence intervals for gestational ages). 138
Pregnancy Hypertension 15 (2019) 134–140
C.J. Petry et al.
Cambridge Baby Growth Study [32], recalled AAM tends to be fairly accurate when compared to AAMs from medical notes [34] (these not being available for this cohort). A third limitation is that missing data meant that numbers of participants varied in the different statistical models, which is not uncommon for cohort studies such as this which was not specifically set up for this particular study. However we overcome potential biases resulting from missing data by only analysing data from women for whom we had at least four blood pressure measurements in pregnancy, and then when using adjustments to check for attenuation we repeated the initial analyses in just those women for whom data relating to the potential confounder was available. Despite this procedure being imperfect, and possibly introducing biases into the population analysed, our data has suggested that blood pressure in pregnancy is robustly related to AAM and that adiposity and insulin resistance may be involved in mediating these associations, which is biologically plausible. In conclusion this study has found negative associations between AAM and subsequent blood pressure in pregnancy (that were not of sufficient effect sizes to change risk of HDP, although positive associations were found between AAM and risk of subsequent gestational hypotension). The associations with blood pressure were attenuated by both BMI and insulin resistance, suggesting that either they may mediate the associations or that there may be a confounder that links AAM to each of them.
[2]
[3]
[4]
[5]
[6]
[7]
Acknowledgments [8]
We thank Catherine Fullah, Nuria Sanz Marcos, Gracielle Pimentel and Niamh Campbell and for help in recording the blood pressure measurements of Cambridge Baby Growth Study participants from the hospital notes. We thank all the families that took part in the Cambridge Baby Growth Study, and we acknowledge the crucial role played by the research nurses especially Suzanne Smith, Ann-Marie Wardell and Karen Forbes, staff at the Addenbrooke’s Wellcome Trust Clinical Research Facility, and midwives at the Rosie Maternity Hospital. Funding for this study has come from the Evelyn Trust (EW9035322), Diabetes UK (11/0004241) and the Wellbeing of Women (the Royal College of Obstetricians and Gynaecologists, UK) (RG1644). Other core funding has come from the Medical Research Council (7500001180), European Union Framework 5 (QLK4-1999-01422), the Mothercare Group Foundation (RG54608), Newlife - The Charity for Disabled Children (07/20), and the World Cancer Research Fund International (2004/03). In addition, there has been support from National Institute for Health Research Cambridge Biomedical Research Centre. The sponsors did not have any role in the study design, in the collection, analysis or the interpretation of the data, in the writing of the manuscript or in the decision to submit it for publication.
[9]
[10]
[11]
[12]
[13]
[14]
Appendix A. Supplementary data
[15]
Supplementary data to this article can be found online at https:// doi.org/10.1016/j.preghy.2019.01.004.
[16]
References
[17]
[1] F.J. Couch, D. Couper, A.D. Coveillo, A. Cox, K. Czene, A.P. D'adamo, G.D. Smith, I. De Vivo, E.W. Demerath, J. Dennis, P. Devilee, A.K. Dieffenbach, A.M. Dunning, G. Eiriksdottir, J.G. Eriksson, P.A. Fasching, L. Ferrucci, D. Flesch-Janys, H. Flyger, T. Foroud, L. Franke, M.E. Garcia, M. García-Closas, F. Geller, E.E. de Geus, G.G. Giles, D.F. Gudbjartsson, V. Gudnason, P. Guénel, S. Guo, P. Hall, U. Hamann, R. Haring, C.A. Hartman, A.C. Heath, A. Hofman, M.J. Hooning, J.L. Hopper, F.B. Hu, D.J. Hunter, D. Karasik, D.P. Kiel, J.A. Knight, V.M. Kosma, Z. Kutalik, S. Lai, D. Lambrechts, A. Lindblom, R. Mägi, P.K. Magnusson, A. Mannermaa, N.G. Martin, G. Masson, P.F. McArdle, W.L. McArdle, M. Melbye, K. Michailidou, E. Mihailov, L. Milani, R.L. Milne, H. Nevanlinna, P. Neven, E.A. Nohr, A.J. Oldehinkel, B.A. Oostra, A. Palotie, M. Peacock, N.L. Pedersen, P. Peterlongo, J. Peto, P.D. Pharoah, D.S. Postma, A. Pouta, K. Pylkäs, P. Radice, S. Ring, F. Rivadeneira, A. Robino, L.M. Rose, A. Rudolph, V. Salomaa, S. Sanna, D. Schlessinger, M.K. Schmidt, M.C. Southey, U. Sovio, M.J. Stampfer, D. Stöckl, A.M. Storniolo, N.J. Timpson, J. Tyrer,
[18]
[19]
[20]
[21]
139
J.A. Visser, P. Vollenweider, H. Völzke, G. Waeber, M. Waldenberger, H. Wallaschofski, Q. Wang, G. Willemsen, R. Winqvist, B.H. Wolffenbuttel, M.J. Wright; Australian Ovarian Cancer Study; GENICA Network; kConFab; LifeLines Cohort Study; InterAct Consortium; Early Growth Genetics (EGG) Consortium, D.I. Boomsma, M.J. Econs, K.T. Khaw, R.J. Loos, M.I. McCarthy, G.W. Montgomery, J.P. Rice, E.A. Streeten, U. Thorsteinsdottir, C.M. van Duijn, B.Z. Alizadeh, S. Bergmann, E. Boerwinkle, H.A. Boyd, L. Crisponi, P. Gasparini, C. Gieger, T.B. Harris, E. Ingelsson, M.R. Järvelin, P. Kraft, D. Lawlor, A. Metspalu, C.E. Pennell, P.M. Ridker, H. Snieder, T.I. Sørensen, T.D. Spector, D.P. Strachan, A.G. Uitterlinden, N.J. Wareham, E. Widen, M. Zygmunt, A. Murray, D.F. Easton, K. Stefansson, J.M. Murabito, K.K. Ong, Parent-of-origin-specific allelic associations among 106 genomic loci for age at menarche, Nature 514 (2014) 92–97. https://www.nature. com/articles/nature13545. A. Yermachenko, V. Dvornyk, Nongenetic determinants of age at menarche: a systematic review, Biomed. Res. Int. 2014 (2014) 371583https://www.hindawi. com/journals/bmri/2014/371583/. J. Luijken, Y.T. van der Schouw, D. Mensink, N.C. Onland-Moret, Association between age at menarche and cardiovascular disease: A systematic review on risk and potential mechanisms, Maturitas 104 (2017) 96–116 https://www.sciencedirect. com/science/article/pii/S0378512217307375?via%3Dihub. M. Janghorbani, M. Mansourian, E. Hosseini, Systematic review and meta-analysis of age at menarche and risk of type 2 diabetes, Acta Diabetol. 51 (2014) 519–528 https://link.springer.com/article/10.1007%2Fs00592-014-0579-x. D. Apter, R. Vihko, Premenarcheal endocrine changes in relation to age at menarche, Clin. Endocrinol. (Oxf) 22 (1985) 753–760 https://onlinelibrary.wiley. com/doi/abs/10.1111/j.1365-2265.1985.tb00165.x. C.J. Petry, K.K. Ong, D.B. Dunger, Age at menarche and the future risk of gestational diabetes: a systematic review and dose response meta-analysis, Acta Diabetol. 55 (2018) 1209–1219 http://www.doi.org/10.1007/s00592-018-1214-z. N. Sudo, T. Shimizu, C. Kawashima, E. Kaneko, M. Tetsuka, A. Miyamoto, Insulinlike growth factor-I (IGF-I) system during follicle development in the bovine ovary: relationship among IGF-I, type 1 IGF receptor (IGFR-1) and pregnancy-associated plasma protein-A (PAPP-A), Mol. Cell. Endocrinol. 264 (2007) 197–203 https:// www.sciencedirect.com/science/article/pii/S0303720706004977?via%3Dihub. C.J. Petry, K.K. Ong, I.A. Hughes, C.L. Acerini, D.B. Dunger, The association between age at menarche and later risk of gestational diabetes is mediated by insulin resistance, Acta Diabetol. 55 (2018) 853–859 https://link.springer.com/article/10. 1007%2Fs00592-018-1162-7. C.J. Petry, K.K. Ong, I.A. Hughes, C.L. Acerini, J. Frystyk, D.B. Dunger, Early pregnancy-associated plasma protein A concentrations are associated with third trimester insulin sensitivity, J. Clin. Endocrinol. Metab. 102 (2017) 2000–2008 https://academic.oup.com/jcem/article/102/6/2000/3067658. G. Valdés, Preeclampsia and cardiovascular disease: interconnected paths that enable detection of the subclinical stages of obstetric and cardiovascular diseases, Integr. Blood Press. Control 10 (2017) 17–23 https://www.dovepress.com/ preeclampsia-and-cardiovascular-disease-interconnected-paths-that-enab-peerreviewed-article-IBPC. E. Parretti, A. Lapolla, M.G. Dalfra, G. Pacini, A. Mari, R. Cioni, C. Marzari, G. Scarselli, G. Mello, Preeclampsia in lean normotensive normotolerant pregnant women can be predicted by simple insulin sensitivity indexes, Hypertension 47 (2006) 449–453 https://www.ahajournals.org/doi/10.1161/01.HYP.0000205122. 47333.7f. D.F. Abetew D.A. Enquobahrie M. Dishi C.B. Rudra R.S. Miller M.A. Williams Age at menarche, menstrual characteristics, and risk of preeclampsia, I.S.R.N. Obstet. Gynecol. 2011, 472083 https://www.hindawi.com/journals/isrn/2011/472083/. A. Anyikam, E. Triche, J. Moodley, E. Norwitz, Multiple sexual partners and early menarche are risk factors for preeclampsia in an African population, Am. J. Obs. Gynec. 191 (2004) S43 https://www.ajog.org/article/S0002-9378(04)01185-8/ abstract. C.L. Rudra, M.A. Williams, BMI as a modifying factor in the relations between age at menarche, menstrual cycle characteristics, and risk of preeclampsia, Gynecol. Endocrinol. 21 (2005) 200–205 https://www.tandfonline.com/doi/abs/10.1080/ 09513590500168431?journalCode=igye20. R.K.S. Gandhi, V. Rao, Socio-demographic and other risk factors of pre-eclampsia at a tertiary care hospital, Karnataka: case control study, J. Clin. Diagn. Res. 8 (2014) JC01-JC04 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4225915/. F.R. Day, C.E. Elks, A. Murray, K.K. Ong, J.R. Perry, Puberty timing associated with diabetes, cardiovascular disease and also diverse health outcomes in men and women: the UK Biobank study, Sci. Rep. 5 (2015) 11208 https://www.nature.com/ articles/srep11208. J.C. Levy, D.R. Matthews, M.P. Hermans, Correct homeostasis model assessment (HOMA) evaluation uses the computer program, Diabetes Care 21 (1998) 2191–2192 http://care.diabetesjournals.org/content/21/12/2191. C.J. Petry, N. Sanz Marcos, G. Pimentel, M.G. Hayes, M. Nodzenski, D.M. Scholtens, I.A. Hughes, C.L. Acerini, K.K. Ong, W.L. Jr, D.B. Dunger Lowe, Associations between fetal imprinted genes and maternal blood pressure in pregnancy, Hypertension. 68 (2016) 1459–1466 https://www.ahajournals.org/doi/10.1161/ HYPERTENSIONAHA.116.08261. National Institute for Health and Care Excellence, Hypertension in pregnancy: diagnosis and management. NICE guidelines [CG107]. August 2010, http://www. nice.org.uk/guidance/cg107 (accessed 28 July, 2016). B. de Lauzon-Guillain, K. Wijndaele, M. Clark, C.L. Acerini, I.A. Hughes, D.B. Dunger, J.C. Wells, K.K. Ong, Breastfeeding and infant temperament at age three months, P.L.o.S, One 7 (2012) e29326http://journals.plos.org/plosone/ article?id=10.1371/journal.pone.0029326. P.M. Catalano, Obesity, insulin resistance, and pregnancy outcome, Reproduction
Pregnancy Hypertension 15 (2019) 134–140
C.J. Petry et al.
[22]
[23]
[24]
[25]
[26]
[27]
[28]
140 (2010) 365–371 http://www.reproduction-online.org/content/140/3/365. long. C.N. Lumeng, A.R. Saltiel, Inflammatory links between obesity and metabolic disease, J. Clin. Invest. 121 (2011) 2111–2117 https://www.jci.org/articles/view/ 57132. H. Sell, C. Habich, J. Eckel, Adaptive immunity in obesity and insulin resistance, Nat. Rev. Endocrinol. 8 (2012) 709–716 https://www.nature.com/articles/nrendo. 2012.114. M.S. Zhou, A. Wang, H. Yu, Link between insulin resistance and hypertension: What is the evidence from evolutionary biology? Diabetol. Metab. Syndr. 6 (2014) 12 https://dmsjournal.biomedcentral.com/articles/10.1186/1758-5996-6-12. K.B. Clancy, L.D. Klein, A. Ziomkiewicz, I. Nenko, G. Jasienska, R.G. Bribiescas, Relationships between biomarkers of inflammation, ovarian steroids, and age at menarche in a rural Polish sample, Am. J. Hum. Biol. 25 (2013) 389–398 https:// onlinelibrary.wiley.com/doi/abs/10.1002/ajhb.22386. F. Juul, V.W. Chang, P. Brar, N. Parekh, Birth weight, early life weight gain and age at menarche: a systematic review of longitudinal studies, Obes. Rev. 18 (2017) 1272–1288 https://onlinelibrary.wiley.com/doi/abs/10.1111/obr.12587. L.G. Grunnet, S. Hansen, L. Hjort, C.M. Madsen, F.B. Kampmann, A.C.B. Thuesen, C. Granstrømi, M. Strøm, E. Maslova, R. Frikke-Schmidt, P. Damm, J.E. Chavarro, F.B. Hu, S.F. Olsen, A. Vaag, Adiposity, dysmetabolic traits, and earlier onset of female puberty in adolescent offspring of women with gestational diabetes mellitus: a clinical study within the Danish National Birth Cohort, Diabetes Care 40 (2017) 1746–1755 http://care.diabetesjournals.org/content/40/12/1746.long. I. Bralić, H. Tahirović, D. Matanić, O. Vrdoljak, S. Stojanović-Spehar, V. Kovacić,
[29]
[30]
[31]
[32]
[33]
[34]
140
S. Blazeković-Milaković, Association of early menarche age and overweight/obesity, J. Pediatr. Endocrinol. Metab. 25 (2012) 57–62 https://www.degruyter.com/ view/j/jpem.2012.25.issue-1-2/jpem-2011-0277/jpem-2011-0277.xml. K.S. Shim, Pubertal growth and epiphyseal fusion, Ann. Pediatr. Endocrinol. Metab. 20 (2015) 8–12 https://e-apem.org/journal/view.php?doi=10.6065/apem.2015. 20.1.8. R. Lakshman, N.G. Forouhi, S.J. Sharp, R. Luben, S.A. Bingham, K.T. Khaw, N.J. Wareham, K.K. Ong, Early age at menarche associated with cardiovascular disease and mortality, J. Clin. Endocrinol. Metab. 94 (2009) 4953–4960 https:// academic.oup.com/jcem/article/94/12/4953/2597000. K.K. Ong, D.B. Dunger, Perinatal growth failure: the road to obesity, insulin resistance and cardiovascular disease in adults, Best Pract. Res. Clin. Endocrinol. Metab. 16 (2002) 191–207 https://www.sciencedirect.com/science/article/pii/ S1521690X02901958?via%3Dihub. P. Prentice, C.L. Acerini, A. Eleftheriou, I.A. Hughes, K.K. Ong, D.B. Dunger, Cohort profile: the Cambridge Baby Growth Study (CBGS), Int. J. Epidemiol. 45 (2016) 35.a-g https://academic.oup.com/ije/article/45/1/35/236371. K.K. Ong, M.L. Ahmed, D.B. Dunger, Lessons from large population studies on timing and tempo of puberty (secular trends and relation to body size): the European trend, Mol. Cell. Endocrinol. 254–255 (2006) 8–12 https://www. sciencedirect.com/science/article/pii/S030372070600219X?via%3Dihub. R. Cooper, M. Blell, R. Hardy, S. Black, T.M. Pollard, M.E. Wadsworth, M.S. Pearce, D. Kuh, Validity of age at menarche self-reported in adulthood, J. Epidemiol. Community Health 60 (2006) 993–997 https://jech.bmj.com/content/60/11/993. long.
Contents lists available at ScienceDirect
Pregnancy Hypertension journal homepage: www.elsevier.com/locate/preghy
Age at menarche and blood pressure in pregnancy a,⁎
a,b,c
Clive J. Petry , Ken K. Ong a b c
a
T a
, Ieuan A. Hughes , Carlo L. Acerini , David B. Dunger
a,c
Department of Paediatrics, University of Cambridge, Cambridge, UK Medical Research Council Epidemiology Unit, University of Cambridge, Cambridge, UK The Institute of Metabolic Science, University of Cambridge, Cambridge, UK
A R T I C LE I N FO
A B S T R A C T
Keywords: Gestational hypertension Gestational hypotension Pre-eclampsia Obesity Development
Objectives: To investigate whether age at menarche is related to maternal blood pressure in pregnancy and, if so, whether obesity and insulin resistance can modify the associations. Study design: Analysis of data collected from 438 pregnant women from the longitudinal and prospective Cambridge Baby Growth Study. Main outcome: Testing associations between questionnaire-derived age at menarche and blood pressure measurements in pregnancy collected from hospital notes, and investigating whether any associations were altered by maternal pre-pregnancy body mass index (BMI) and insulin resistance. Measures: Mean arterial blood pressure at four time points across pregnancy, age at menarche, (Homeostasis Model Assessment) insulin resistance around week 28 of pregnancy. Results: For each increased year in age at menarche there was a drop in mean arterial blood pressure (mmHg) of 0.6 at 11.9 weeks, 0.9 at 31.4 and 37.0 weeks, and 0.4 at 38.8 weeks (a maximal difference of over 7 mmHg across extremes of AAM). Each association was attenuated by both maternal pre-pregnancy BMI and insulin resistance. Conclusions: Age at menarche is negatively associated with future blood pressure in pregnancy, so those with the earliest age at menarche have the highest blood pressures. Either these associations may be mediated by links between age at menarche and obesity/insulin resistance, or there may be a confounder (e.g. systemic inflammation) that links age at menarche to each of them.
1. Introduction Age at menarche (AAM) is influenced by a wide range of genetic [1] and non-genetic determinants [2]. Given that some of the genetic variants associated with AAM are also associated with body mass index (BMI) it is perhaps not surprising that AAM is also associated with some BMI-related pathological conditions such as cardiovascular disease [3] and type 2 diabetes [4]. Early AAM, which accompanies most of the increased risk for these conditions, appears to result from increased follicle-stimulating hormone (FSH) and oestradiol concentrations [5], the latter appearing to be continually raised above expected levels into puberty and beyond possibly because of a reduced sensitivity of the hypothalamic-pituitary axis to the negative feedback of circulating steroid hormones. The fact that AAM is also related to the risk of developing another BMI-linked condition in pregnancy, gestational diabetes [6], raises the possibility that the hormonal changes involved in altering AAM may also affect pregnancy hormone concentrations. Indeed, there is in vitro
⁎
evidence from cultured bovine granulosa cells that FSH is able to alter expression of pregnancy-associated plasma protein A (PAPP-A) [7]. We recently found non-linear associations between AAM and serum PAPPA concentrations [8], which could be important because of the potential role of PAPP-A in regulating pregnancy insulin sensitivity [9]. In our study AAM was negatively associated with insulin resistance in pregnancy [8]. Given the links between AAM and cardiovascular disease [3], pre-eclampsia and cardiovascular disease [10] and pre-eclampsia with insulin resistance [11], it is not surprising that AAM has been found to be negatively associated with pre-eclampsia in most [12–15] although not all [16] studies. The first sign of pre-eclampsia is often raised blood pressure. Although AAM is associated with various cardiovascular traits outside of pregnancy its association with blood pressure in pregnancy has not been thoroughly investigated. This study was therefore implemented to investigate potential associations between AAM and blood pressure across pregnancy.
Corresponding author at: Department of Paediatrics, Box 116, Cambridge Biomedical Campus, Hills Road, Cambridge CB2 0QQ, UK. E-mail address: [email protected] (C.J. Petry).
https://doi.org/10.1016/j.preghy.2019.01.004 Received 21 September 2018; Received in revised form 11 December 2018; Accepted 10 January 2019 Available online 11 January 2019 2210-7789/ © 2019 International Society for the Study of Hypertension in Pregnancy. Published by Elsevier B.V. All rights reserved.
Pregnancy Hypertension 15 (2019) 134–140
C.J. Petry et al.
2. Materials and methods
guidelines. Written informed consent was obtained from all the study participants.
2.1. Cohort 2.6. Assays Between 2001 and 2009 the prospective and longitudinal Cambridge Baby Growth Study recruited 2229 mothers (and their and offspring) attending ultrasound clinics during early pregnancy at the Rosie Maternity Hospital, Cambridge, United Kingdom [9]. The study participants, all of whom were over 16 years of age and none of whom reported suffering from chronic hypertension or were treated with antihypertensives prior to conception, received routine care during pregnancy including the monitoring of their blood pressure. In addition fasting blood samples were collected from 1239 participants (55.6%) for the measurement of plasma glucose and insulin concentrations around week 28 of pregnancy for the evaluation of insulin resistance by the Homeostasis Model Assessment (HOMA) [17]. In this cohort, 96.9% of the offspring were of white ethnicity, 0.9% were Indo-Asian, 0.8% were East-Asian, 0.6% were black (African or Caribbean) and a further 0.8% were of mixed race.
PAPP-A was measured by time-resolved fluoroimmunoassay (autoDELFIA, Perkin Elmer Ltd., Seer Green, U.K.). The minimum detection limit of this assay was 5 mIU/L. The intra- and inter-assay coefficients of variation (CV) was less than 8% and 10% throughout. Insulin concentrations were measured by enzyme-linked immunosorbent assay using a commercial kit (DSL, London, U.K.). The minimum detection limit of this assay was 1.6 pmol/L. Intra-assay CVs were 4.4 and 5.1% at 62 pmol/L and 215 pmol/L, and equivalent interassay CVs were 8.7 and 2.9%; this assay has no cross-reactivity with proinsulin at levels up to 1,000 pmol/L. Blood glucose concentrations were measured using a routine glucose oxidase-based method. 2.7. Calculations The BMI before pregnancy was calculated as the pre-pregnancy body weight divided by the height squared. The reproductive age was determined as the maternal age at the time of the birth of her baby in the index pregnancy minus the AAM. Mean arterial blood pressure in pregnancy was calculated as the sum of twice the diastolic plus the systolic blood pressure, divided by three. HOMA IR was estimated from the fasting glucose and insulin concentrations using the HOMA calculator available at https://www.dtu.ox.ac.uk/homacalculator/ [17]. The index of multiple deprivation was derived from residential post codes as described [20].
2.2. Age at menarche Each of the study participants was given a printed questionnaire at recruitment to complete during the rest of the pregnancy [8], of which a total of 1273 women (57.1%) did so. One of the questions asked “What age were you when you had your first period?” The mean (95% confidence interval) AAM of those women was 12.9 (12.9, 13.0) years. 2.3. Blood pressure during pregnancy
2.8. Statistical analysis
Records of routine blood pressure measurements made during pregnancy were collected from the hospital notes from 968 women in the Cambridge Baby Growth Study (hospital notes for the other study participants either not being available or not containing record of these blood pressures) [18]. Of all the readings four times in pregnancy emerged as points when blood pressure was routinely measured in the majority of subjects. According to the gestational week at which the measurements were taken, the first three time points were: (1) at 11.8 (11.5, 12.0) weeks, (2) at 31.4 (31.3, 31.5) weeks and (3) at 37.0 (36.9, 37.0) weeks. The fourth readings were taken specifically during the final 2 weeks prior to parturition (mean 38.8 (38.7, 38.9) weeks). The present analysis was restricted to data from 438 pregnancies where all four blood pressure measurements and the AAM were available.
Associations between (mean arterial, systolic or diastolic) blood pressure and AAM were assessed by linear regression (adjusted only for the exact stage of pregnancy or both that and either maternal prepregnancy BMI or week 28 HOMA IR). Associations with serial blood pressure measurements were analysed by general estimation equation modelling, adjusting for weeks of gestation ( ± BMI or HOMA IR) when the blood pressure readings were taken. Associations between other continuous variables (adjusted for confounders where appropriate) were also assessed by linear regression whereas associations with HDP or gestational hypotension were assessed using logistic regression. Associations between categorical variables were tested using the χ2test. Unless otherwise stated all other data are presented as means (95% confidence intervals). Statistical analyses were performed using Stata 13 (StataCorp LP, College Station, Texas, U.S.A.). P < 0.05 was considered statistically significant throughout.
2.4. Gestational hypertension & hypotension When hospital notes were investigated additional evidence of hypertensive disorders of pregnancy (HDP) was also sought (defined using the inclusion of a diagnosis of the following terms: “pre-eclampsia”, “gestational hypertension” or “pregnancy-induced hypertension”). In addition further women were categorised as having HDP using the National Institute for Health and Care Excellence criteria for gestational hypertension (blood pressure measurements in the second half of pregnancy that are > = 140 mmHg systolic or 90 mmHg diastolic blood pressure in women without chronic hypertension) [19], with the exception that for our study evidence of HDP was accepted if the blood pressure cut offs were exceeded at one reading rather than at least two [18]. Evidence of gestational hypotension was defined in this study as any one of the blood pressure recordings for a study participant being < = 110/60 mmHg.
3. Results 3.1. Study participants There were no differences between those from the Cambridge Baby Growth Study that were included in the present analysis and those that were not in terms of maternal age, pre-pregnancy BMI, blood pressures in the second half of pregnancy, week 15 PAPP-A concentrations, week 28 glucose and insulin concentrations and HOMA IR, and the baby’s birth weight adjusted for known confounders (Table 1). However fewer of those included in the present analysis smoked in pregnancy and they had lower blood pressures at booking clinic, higher gestational ages at the birth of their babies (of an average of around 4 days) and higher unadjusted baby birth weights as well as lower indexes of multiple deprivation. A higher proportion were also nulliparous and developed HDP, although there was no difference in the proportion that developed pre-eclampsia. Mean arterial blood pressure at booking clinic was negatively associated with AAM (see below) in the women included in the
2.5. Ethics The Cambridge Baby Growth Study was approved by the local ethics committee, Addenbrooke’s Hospital, Cambridge, United Kingdom. All procedures followed were in accordance with the institutional 135
Pregnancy Hypertension 15 (2019) 134–140
3.2. Associations between AAM and blood pressure in pregnancy At booking clinic AAM showed a negative association with mean arterial blood pressure (Table 2). By weeks 31 and 37 the effect sizes of the negative associations were greater in terms of effect size and significance. Just prior to the birth of their baby the effect size of the 136
0.5 −0.3 (–1.1, +0.6) 308 0.2 −0.5 (−1.4, +0.3) 0.9 0.3
402
−0.1 (–0.6, +0.8)
308
0.2 0.06 −0.4 (–1.1, +0.3) −0.7 (–1.5, 0) 308 308 0.07 9.6 × 10−3 −0.7 (–1.4, 0) −1.1 (–1.8, –0.3) 308 308 0.1 0.1 −0.6 (–1.0, +0.1) −0.5 (–1.1, +0.2) 402 402 2.0 × 10−3 4.2 × 10−3
−0.1 (–0.8, +0.6) 308 0.3 −0.3 (–1.1, +0.4) 308 0.7 −0.1 (–0.7, +0.5) 402 0.07
Change in Mean Arterial Blood Pressure (mmHg/year) N N p-value
402 0.2
Data are means (95% confidence interval). Model 1 is adjusted for gestational age when the blood pressures were measured. Model 2 is adjusted for gestational age when the blood pressures were measured in women with pre-pregnancy BMI recorded. Model 3 is adjusted for gestational age and pre-pregnancy BMI when the blood pressures were measured in women with BMI recorded. Model 4 is adjusted for gestational age when the blood pressures were measured in women with week 28 HOMA IR recorded. Model 5 is adjusted for gestational age and week 28 HOMA IR when the blood pressures were measured in women with HOMA IR recorded.
present analysis, but there was no significant association of AAM with smoking in pregnancy (p = 0.2), gestational age at the birth of their babies (p = 1.0), unadjusted birth weights (p = 0.2), index of multiple deprivation (p = 0.2) or parity (p = 0.08).
438
Data are means (95% confidence intervals). *adjusted for gestational age at birth, sex, parity and maternal pre-pregnancy BMI.
Change in Mean Arterial Blood Pressure (mmHg/year)
p-value
0.4
38.8 (38.7, 38.9)
3.469 (3.438, 3.500) (n = 779)
402 402
Baby’s adjusted birth weight (kg)*
5.1 × 10–5
1.7 × 10−3 7.1 × 10−3
Baby’s unadjusted birth weight (kg)
3.6 × 10–10
438 438
Gestational age of baby at birth (weeks)
0.5
31.4 (31.3, 31.5) 37.0 (36.9, 37.1)
Week 28 HOMA IR (%)
0.3
−0.6 (–1.2, 0) −1.0 (–1.6, –0.4) −0.9 (−1.6, −0.3) −0.4 (−1.1, +0.3)
Week 28 fasting insulin concentration (pmol/L)
0.7
402
Week 28 fasting glucose concentration (mmol/L)
0.3
0.05
Week 15 serum PAPP-A concentration (mIU/L)
0.5
−0.6 (–1.2, 0) −0.9 (–1.3, –0.3) −0.9 (−1.5, −0.2) −0.4 (−1.1, +0.3)
Pre-eclampsia (n)
0.01
438
HDP (n)
0.8
11.9 (11.6, 12.1)
Mean arterial blood pressure (mmHg) at 38.8 weeks
0.6
Change in Mean Arterial Blood Pressure (mmHg/year)
Mean arterial blood pressure (mmHg) at 37.0 weeks
0.3
p-value
Mean arterial blood pressure (mmHg) at 31.4 weeks
0.02
Model 4
Mean arterial blood pressure (mmHg) at 11.7 weeks
6.6 × 10–3
Model 3
Index of multiple deprivation
0.03
Model 2
Smoked during pregnancy (n)
0.01
Model 1
Parity (n nulliparous)
0.3
Table 2 Associations between AAM and mean arterial blood pressure in pregnancy in the Cambridge Baby Growth Study in women for whom all data were available.
Pre-pregnancy BMI (kg/m2)
N
0.5
Change in Mean Arterial Blood Pressure (mmHg/year)
33.6 (33.3, 33.8) (n = 908) 24.2 (23.9, 24.5) (n = 785) 503 out of 1208 (41.6%) 72 out of 1218 (5.9%) 11.1 (9.7, 12.6) (n = 40) 81.7 (80.7, 82.8) (n = 260) 82.0 (81.0, 83.1) (n = 265) 85.2 (84.0, 86.4) (n = 247) 86.6 (84.3, 88.9) (n = 90) 19 out of 530 (3.6%) 12 out of 530 (2.3%) 6879 (6331, 7427) (n = 582) 4.3 (4.3, 4.4) (n = 770) 53 (50, 56) (n = 806) 93 (89, 96) (n = 749) 39.6 (39.5, 39.7) (n = 1,219) 3.448 (3.417, 3.478) (n = 1,213)
N
33.4 (33.0, 33.8) (n = 427) 23.9 (23.4, 24.3) (n = 402) 214 out of 438 (48.9%) 14 out of 437 (3.2%) 9.0 (8.5, 9.5) (n = 321) 80.1 (79.2, 80.9) (n = 438) 81.4 (80.6, 82.2) (n = 438) 85.0 (83.9, 85.7) (n = 438) 87.0 (86.0, 88.0) (n = 438) 32 out of 438 (7.3%) 7 out of 438 (1.6%) 6999 (6143, 7854) (n = 239) 4.3 (4.3, 4.4) (n = 313) 52 (48, 56) (n = 330) 90 (85, 96) (n = 310) 40.2 (40.0, 40.3) (n = 438) 3.563 (3.513, 3.614) (n = 436) 3.494 (3.451, 3.538) (n = 401)
p-value
Maternal age (years)
Change in Mean Arterial Blood Pressure (mmHg/year)
p-value
N
Excluded women
Gestational Age (Weeks)
Included women
Model 5
Characteristic
p-value
Table 1 Characteristics of those Cambridge Baby Growth Study participants who were part of this analysis (from whom we had self-reported AAM and at least 4 blood pressures in pregnancy recorded from hospital notes) and comparison with those participants that were not included in the analysis.
0.7
C.J. Petry et al.
Pregnancy Hypertension 15 (2019) 134–140
C.J. Petry et al.
Table 3 Associations between serial mean arterial blood pressures (4 measurements taken over pregnancy) and AAM as analysed by general estimation equation modelling. Mean Arterial Blood Pressure Effect Size (mmHg/year)
p-value
n
Confounders in model
−0.7 −0.6 −0.7 −0.2 −0.6 −0.4
6.3 × 10−3 7.9 × 10−3 7.1 × 10−3 0.4 0.05 0.2
438 438 402 402 308 308
Gestational Gestational Gestational Gestational Gestational Gestational
(−1.2, (−1.0, (−1.2, (−0.7, (−1.2, (−0.9,
−0.2) −0.2) −0.2) 0.3) 0) 0.2)
age age age age age age
when when when when when when
blood blood blood blood blood blood
pressures pressures pressures pressures pressures pressures
measured, measured, measured, measured, measured, measured,
identity identity identity identity identity identity
number number, HDP, pre-eclampsia number (if BMI data available) number, BMI number (if week 28 HOMA IR data available) number, week 28 HOMA IR
Data are mean (95% confidence interval).
Fig. 1. Predicted systolic blood pressures (after adjustment for gestational age) against AAM in linear regression models from the Cambridge Baby Growth Study: at 11.9 weeks (top left, β = −1.0 (−1.9, −0.2) mmHg/year, p = 0.02), 31.4 weeks (top right, β = −1.1 (–1.9, –0.3) mmHg/year, p = 9.2 × 10−3), 37.0 weeks (bottom left, β = −1.2 (−2.0, −0.4) mmHg/year, p = 5.3 × 10−3) and 38.8 weeks (bottom right, β = −0.8 (−1.6, +0.1) mmHg/year, p = 0.08). N = 438 throughout. Solid lines are the means and the dashed lines are the 95% confidence intervals.
n = 427).
association was smaller (and no longer statistically significant), albeit in the same direction as the earlier associations in pregnancy. Overall using repeat measures there was a significant negative association between AAM and mean arterial blood pressure (Table 3), which was attenuated by both pre-pregnancy BMI and HOMA IR as were the individual associations (Table 2). The relationship between blood pressure and AAM appeared to be stronger with systolic (Fig. 1) rather than diastolic blood pressure (Supplementary Table 1). The reproductive age was not associated with mean arterial blood pressure at any of the four blood pressure readings: 11.9 weeks (β = 0.0 (-0.2, 0.2) mmHg/year, p = 0.9, n = 427), 31.4 weeks (β = -0.1 (-0.3, 0) mmHg/year, p = 0.1, n = 427), 37.0 weeks (β = -0.1 (-0.3, +0.1) mmHg/year, p = 0.3, n = 427), 38.8 weeks (β = -0.2 (-0.4, 0) mmHg/year, p = 0.08,
3.3. Associations between AAM, HDP and gestational hypotension There was no significant association between AAM and HDP (odds ratio (OR) 0.89 (0.67, 1.19) per year, p = 0.5, n = 438), even though they were associated with both HOMA IR (OR 1.6 (1.1, 2.3) per HOMA IR unit, p = 0.03, n = 335) and BMI (OR 1.2 (1.1, 1.3) per kg/m2, p = 5.4 × 10–5, n = 414) in these women. There was also no significant difference in AAM between those that had HDP and those that did not: no HDP 13.0 (12.8, 13.1) years (n = 413) v. HDP 12.8 (12.2, 13.3) years (n = 25) (p = 0.5). There was a significant positive association between AAM and 137
Pregnancy Hypertension 15 (2019) 134–140
C.J. Petry et al.
adjusted for either pre-pregnancy BMI or HOMA IR (both pre-pregnancy BMI and HOMA IR [8] themselves being strongly negatively associated with AAM), suggesting a potential mechanism involving increased adiposity and insulin resistance underpinning the link between AAM and blood pressure in pregnancy. Insulin resistance and obesity are well known associates of increased blood pressure both in and outside of pregnancy [21]. Evolutionary biology suggests that the insulin resistance may not be causal in this relationship however. Rather both it and increased blood pressure may result from the low grade systemic inflammation occurring as a result of the increased macrophage infiltration of adipose tissue [22] and other tissues containing ectopically-stored lipids [23] that occurs in obesity [24]. Consistent with this AAM has been reported to be negatively associated with circulating C-reactive protein concentrations suggesting a link between early AAM and inflammation [25]. Such effects appear to arise from the interaction between historically beneficial “thrifty” genotypes (that favour energy storage) and modern diets and lifestyles. In pregnancy placental hormones can be part of these processes. In the Cambridge Baby Growth Study we previously found that low week 15 PAPP-A concentrations were associated with insulin resistance in pregnancy [9] and that women with an early AAM had lower PAPP-A concentrations than women with AAMs closer to the average [8]. In the present analysis week 15 PAPP-A concentrations were negatively associated with all the blood pressure measurements. Further studies are needed to identify whether PAPP-A is causal in these processes or whether the associations observed are confounded, possibly by the insulin resistance resulting from the low grade systemic inflammation described above. The associations between early AAM and raised blood pressure in pregnancy observed in this study could be part of a particular female life course trajectory framework that is “high risk” for poor cardiometabolic health. AAM has been related to each of the steps in the following sequence: the combination of low birth weight and subsequent catch-up growth in infancy [26], or alternatively in utero exposure to gestational diabetes [27], followed by childhood obesity [28], early AAM and subsequent epiphyseal fusion [29], shorter adult height and adult obesity [8], insulin resistance [8], raised blood pressure (the present study; possibly increasing the risk of pre-eclampsia [12–15]) and increased glucose concentrations [8] in pregnancy, cardiovascular disease [3,30] and type 2 diabetes [4] in later life. Although somewhat speculative there is evidence of links between the other factors [31] even if you remove early AAM from the sequence. However these and the present study suggest that AAM should be considered when deliberating over the developmental origins of health and disease. Although the present study has found apparently robust negative associations between AAM and blood pressure in pregnancy it does have a number of limitations. The first of these is that the women tested may not be entirely representative of the population from which they were sampled. The Cambridge Baby Growth Study cohort as a whole is representative of the South Cambridgeshire population of pregnant women, albeit on average the mothers being slightly older and more likely to be nulliparous [32]. However the subpopulation from whom the present analysis was drawn included women whose blood pressure at booking clinic was slightly lower than that of the cohort as a whole, with their baby’s births occurring on average 4 days later than those not included leading to increased unadjusted birth weights of the babies. A lower proportion of the mothers included in the present analysis smoked (although smoking was uncommon as a whole in this population) and they were less deprived in a population not noted for high levels of deprivation. A higher proportion of them were also nulliparous and developed HDP. However the only one of these that was significantly associated with AAM was the booking clinic blood pressure so any biases may have been limited. Another limitation to the study is that the AAM, at a time when AAMs were thought to be declining in the U.K. at a slow rate [33], was self-reported so some of them may have been inaccurate. However it has previously been reported that in a generally well-qualified population, as is a characteristic of the
Table 4 Associations between circulating PAPP-A concentrations from around week 15 of pregnancy and mean arterial blood pressures in those pregnancies. Gestational Age (Weeks)
Change in Mean Arterial Blood Pressure (SDs/year)
N
p-value
11.9 31.4 37.0 38.8
−0.182 −0.188 −0.206 −0.108
261 261 261 261
0.01 0.01 4.9 × 10−3 0.1
(11.6, (31.3, (36.9, (38.7,
12.1) 31.5) 37.1) 38.9)
Data are means (with 95% confidence intervals for gestational ages).
unadjusted evidence of gestational hypotension (OR 1.2 (1.1, 1.4) per year, p = 2.3 × 10−3, n = 438). Those women who experienced hypotension in pregnancy had significantly later AAMs than those that did not: no hypotension 12.7 (12.5, 12.9) years (n = 193) v. hypotension 13.2 (13.0, 13.3) years (n = 245) (p = 2.0 × 10−3). 3.4. Associations between PAPP-A concentrations, HOMA IR and blood pressure in pregnancy Blood pressure in pregnancy was negatively associated with serum PAPP-A concentrations from around week 15 of pregnancy (Table 4), with the first three measurements having statistically significant associations and the association with the late pregnancy measurement having a relatively smaller (not statistically significant) effect size. All the blood pressure measurements were also positively associated with week 28 HOMA IR (Table 5). We previously reported associations between AAM and: (1) pre-pregnancy BMI, (2) week 15 circulating PAPPA concentrations and (3) week 28 HOMA IR in our population [8]. 4. Discussion In this study we have shown negative associations between AAM and four blood pressure measurements in pregnancy. Whilst only three of the associations reached statistical significance, two were strong and all of them plus the combined association were consistently in the same direction with just the effect size altering. The highest blood pressures were consistently in those women with early AAM. The effect sizes were not large enough to independently change the risk for HDP in our population, although AAM was positively associated with evidence of gestational hypotension. The lack of association with HDP or even preeclampsia (data not shown) could help explain why AAM is not associated with pre-eclampsia in every study [16]. However with those women with the earliest AAMs having on average predicted mean arterial blood pressure up to over 7 mmHg higher than those with the latest AAMs in our population, factors related to AAM could clearly be associated with changes in risk in combination with other risk factors. The same does not appear to be true for the time elapsed between the AAM and the mother’s age at the time of the birth of her baby. Associations between AAM and mean arterial blood pressure were slightly weaker in the reduced number of women for whom we had prepregnancy BMIs available, presumably due to the accompanying reduction in statistical power. They were attenuated when they were Table 5 Associations between HOMA IR (fasting insulin resistance) from around week 28 of pregnancy and mean arterial blood pressures in those pregnancies. Gestational Age (Weeks)
Change in Mean Arterial Blood Pressure (SDs/year)
N
p-value
11.9 31.4 37.0 38.8
0.221 0.301 0.314 0.235
335 335 335 335
4.8 × 10−5 1.7 × 10−8 2.7 × 10−9 1.6 × 10−5
(11.6, (31.3, (36.9, (38.7,
12.1) 31.5) 37.1) 38.9)
Data are means (with 95% confidence intervals for gestational ages). 138
Pregnancy Hypertension 15 (2019) 134–140
C.J. Petry et al.
Cambridge Baby Growth Study [32], recalled AAM tends to be fairly accurate when compared to AAMs from medical notes [34] (these not being available for this cohort). A third limitation is that missing data meant that numbers of participants varied in the different statistical models, which is not uncommon for cohort studies such as this which was not specifically set up for this particular study. However we overcome potential biases resulting from missing data by only analysing data from women for whom we had at least four blood pressure measurements in pregnancy, and then when using adjustments to check for attenuation we repeated the initial analyses in just those women for whom data relating to the potential confounder was available. Despite this procedure being imperfect, and possibly introducing biases into the population analysed, our data has suggested that blood pressure in pregnancy is robustly related to AAM and that adiposity and insulin resistance may be involved in mediating these associations, which is biologically plausible. In conclusion this study has found negative associations between AAM and subsequent blood pressure in pregnancy (that were not of sufficient effect sizes to change risk of HDP, although positive associations were found between AAM and risk of subsequent gestational hypotension). The associations with blood pressure were attenuated by both BMI and insulin resistance, suggesting that either they may mediate the associations or that there may be a confounder that links AAM to each of them.
[2]
[3]
[4]
[5]
[6]
[7]
Acknowledgments [8]
We thank Catherine Fullah, Nuria Sanz Marcos, Gracielle Pimentel and Niamh Campbell and for help in recording the blood pressure measurements of Cambridge Baby Growth Study participants from the hospital notes. We thank all the families that took part in the Cambridge Baby Growth Study, and we acknowledge the crucial role played by the research nurses especially Suzanne Smith, Ann-Marie Wardell and Karen Forbes, staff at the Addenbrooke’s Wellcome Trust Clinical Research Facility, and midwives at the Rosie Maternity Hospital. Funding for this study has come from the Evelyn Trust (EW9035322), Diabetes UK (11/0004241) and the Wellbeing of Women (the Royal College of Obstetricians and Gynaecologists, UK) (RG1644). Other core funding has come from the Medical Research Council (7500001180), European Union Framework 5 (QLK4-1999-01422), the Mothercare Group Foundation (RG54608), Newlife - The Charity for Disabled Children (07/20), and the World Cancer Research Fund International (2004/03). In addition, there has been support from National Institute for Health Research Cambridge Biomedical Research Centre. The sponsors did not have any role in the study design, in the collection, analysis or the interpretation of the data, in the writing of the manuscript or in the decision to submit it for publication.
[9]
[10]
[11]
[12]
[13]
[14]
Appendix A. Supplementary data
[15]
Supplementary data to this article can be found online at https:// doi.org/10.1016/j.preghy.2019.01.004.
[16]
References
[17]
[1] F.J. Couch, D. Couper, A.D. Coveillo, A. Cox, K. Czene, A.P. D'adamo, G.D. Smith, I. De Vivo, E.W. Demerath, J. Dennis, P. Devilee, A.K. Dieffenbach, A.M. Dunning, G. Eiriksdottir, J.G. Eriksson, P.A. Fasching, L. Ferrucci, D. Flesch-Janys, H. Flyger, T. Foroud, L. Franke, M.E. Garcia, M. García-Closas, F. Geller, E.E. de Geus, G.G. Giles, D.F. Gudbjartsson, V. Gudnason, P. Guénel, S. Guo, P. Hall, U. Hamann, R. Haring, C.A. Hartman, A.C. Heath, A. Hofman, M.J. Hooning, J.L. Hopper, F.B. Hu, D.J. Hunter, D. Karasik, D.P. Kiel, J.A. Knight, V.M. Kosma, Z. Kutalik, S. Lai, D. Lambrechts, A. Lindblom, R. Mägi, P.K. Magnusson, A. Mannermaa, N.G. Martin, G. Masson, P.F. McArdle, W.L. McArdle, M. Melbye, K. Michailidou, E. Mihailov, L. Milani, R.L. Milne, H. Nevanlinna, P. Neven, E.A. Nohr, A.J. Oldehinkel, B.A. Oostra, A. Palotie, M. Peacock, N.L. Pedersen, P. Peterlongo, J. Peto, P.D. Pharoah, D.S. Postma, A. Pouta, K. Pylkäs, P. Radice, S. Ring, F. Rivadeneira, A. Robino, L.M. Rose, A. Rudolph, V. Salomaa, S. Sanna, D. Schlessinger, M.K. Schmidt, M.C. Southey, U. Sovio, M.J. Stampfer, D. Stöckl, A.M. Storniolo, N.J. Timpson, J. Tyrer,
[18]
[19]
[20]
[21]
139
J.A. Visser, P. Vollenweider, H. Völzke, G. Waeber, M. Waldenberger, H. Wallaschofski, Q. Wang, G. Willemsen, R. Winqvist, B.H. Wolffenbuttel, M.J. Wright; Australian Ovarian Cancer Study; GENICA Network; kConFab; LifeLines Cohort Study; InterAct Consortium; Early Growth Genetics (EGG) Consortium, D.I. Boomsma, M.J. Econs, K.T. Khaw, R.J. Loos, M.I. McCarthy, G.W. Montgomery, J.P. Rice, E.A. Streeten, U. Thorsteinsdottir, C.M. van Duijn, B.Z. Alizadeh, S. Bergmann, E. Boerwinkle, H.A. Boyd, L. Crisponi, P. Gasparini, C. Gieger, T.B. Harris, E. Ingelsson, M.R. Järvelin, P. Kraft, D. Lawlor, A. Metspalu, C.E. Pennell, P.M. Ridker, H. Snieder, T.I. Sørensen, T.D. Spector, D.P. Strachan, A.G. Uitterlinden, N.J. Wareham, E. Widen, M. Zygmunt, A. Murray, D.F. Easton, K. Stefansson, J.M. Murabito, K.K. Ong, Parent-of-origin-specific allelic associations among 106 genomic loci for age at menarche, Nature 514 (2014) 92–97. https://www.nature. com/articles/nature13545. A. Yermachenko, V. Dvornyk, Nongenetic determinants of age at menarche: a systematic review, Biomed. Res. Int. 2014 (2014) 371583https://www.hindawi. com/journals/bmri/2014/371583/. J. Luijken, Y.T. van der Schouw, D. Mensink, N.C. Onland-Moret, Association between age at menarche and cardiovascular disease: A systematic review on risk and potential mechanisms, Maturitas 104 (2017) 96–116 https://www.sciencedirect. com/science/article/pii/S0378512217307375?via%3Dihub. M. Janghorbani, M. Mansourian, E. Hosseini, Systematic review and meta-analysis of age at menarche and risk of type 2 diabetes, Acta Diabetol. 51 (2014) 519–528 https://link.springer.com/article/10.1007%2Fs00592-014-0579-x. D. Apter, R. Vihko, Premenarcheal endocrine changes in relation to age at menarche, Clin. Endocrinol. (Oxf) 22 (1985) 753–760 https://onlinelibrary.wiley. com/doi/abs/10.1111/j.1365-2265.1985.tb00165.x. C.J. Petry, K.K. Ong, D.B. Dunger, Age at menarche and the future risk of gestational diabetes: a systematic review and dose response meta-analysis, Acta Diabetol. 55 (2018) 1209–1219 http://www.doi.org/10.1007/s00592-018-1214-z. N. Sudo, T. Shimizu, C. Kawashima, E. Kaneko, M. Tetsuka, A. Miyamoto, Insulinlike growth factor-I (IGF-I) system during follicle development in the bovine ovary: relationship among IGF-I, type 1 IGF receptor (IGFR-1) and pregnancy-associated plasma protein-A (PAPP-A), Mol. Cell. Endocrinol. 264 (2007) 197–203 https:// www.sciencedirect.com/science/article/pii/S0303720706004977?via%3Dihub. C.J. Petry, K.K. Ong, I.A. Hughes, C.L. Acerini, D.B. Dunger, The association between age at menarche and later risk of gestational diabetes is mediated by insulin resistance, Acta Diabetol. 55 (2018) 853–859 https://link.springer.com/article/10. 1007%2Fs00592-018-1162-7. C.J. Petry, K.K. Ong, I.A. Hughes, C.L. Acerini, J. Frystyk, D.B. Dunger, Early pregnancy-associated plasma protein A concentrations are associated with third trimester insulin sensitivity, J. Clin. Endocrinol. Metab. 102 (2017) 2000–2008 https://academic.oup.com/jcem/article/102/6/2000/3067658. G. Valdés, Preeclampsia and cardiovascular disease: interconnected paths that enable detection of the subclinical stages of obstetric and cardiovascular diseases, Integr. Blood Press. Control 10 (2017) 17–23 https://www.dovepress.com/ preeclampsia-and-cardiovascular-disease-interconnected-paths-that-enab-peerreviewed-article-IBPC. E. Parretti, A. Lapolla, M.G. Dalfra, G. Pacini, A. Mari, R. Cioni, C. Marzari, G. Scarselli, G. Mello, Preeclampsia in lean normotensive normotolerant pregnant women can be predicted by simple insulin sensitivity indexes, Hypertension 47 (2006) 449–453 https://www.ahajournals.org/doi/10.1161/01.HYP.0000205122. 47333.7f. D.F. Abetew D.A. Enquobahrie M. Dishi C.B. Rudra R.S. Miller M.A. Williams Age at menarche, menstrual characteristics, and risk of preeclampsia, I.S.R.N. Obstet. Gynecol. 2011, 472083 https://www.hindawi.com/journals/isrn/2011/472083/. A. Anyikam, E. Triche, J. Moodley, E. Norwitz, Multiple sexual partners and early menarche are risk factors for preeclampsia in an African population, Am. J. Obs. Gynec. 191 (2004) S43 https://www.ajog.org/article/S0002-9378(04)01185-8/ abstract. C.L. Rudra, M.A. Williams, BMI as a modifying factor in the relations between age at menarche, menstrual cycle characteristics, and risk of preeclampsia, Gynecol. Endocrinol. 21 (2005) 200–205 https://www.tandfonline.com/doi/abs/10.1080/ 09513590500168431?journalCode=igye20. R.K.S. Gandhi, V. Rao, Socio-demographic and other risk factors of pre-eclampsia at a tertiary care hospital, Karnataka: case control study, J. Clin. Diagn. Res. 8 (2014) JC01-JC04 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4225915/. F.R. Day, C.E. Elks, A. Murray, K.K. Ong, J.R. Perry, Puberty timing associated with diabetes, cardiovascular disease and also diverse health outcomes in men and women: the UK Biobank study, Sci. Rep. 5 (2015) 11208 https://www.nature.com/ articles/srep11208. J.C. Levy, D.R. Matthews, M.P. Hermans, Correct homeostasis model assessment (HOMA) evaluation uses the computer program, Diabetes Care 21 (1998) 2191–2192 http://care.diabetesjournals.org/content/21/12/2191. C.J. Petry, N. Sanz Marcos, G. Pimentel, M.G. Hayes, M. Nodzenski, D.M. Scholtens, I.A. Hughes, C.L. Acerini, K.K. Ong, W.L. Jr, D.B. Dunger Lowe, Associations between fetal imprinted genes and maternal blood pressure in pregnancy, Hypertension. 68 (2016) 1459–1466 https://www.ahajournals.org/doi/10.1161/ HYPERTENSIONAHA.116.08261. National Institute for Health and Care Excellence, Hypertension in pregnancy: diagnosis and management. NICE guidelines [CG107]. August 2010, http://www. nice.org.uk/guidance/cg107 (accessed 28 July, 2016). B. de Lauzon-Guillain, K. Wijndaele, M. Clark, C.L. Acerini, I.A. Hughes, D.B. Dunger, J.C. Wells, K.K. Ong, Breastfeeding and infant temperament at age three months, P.L.o.S, One 7 (2012) e29326http://journals.plos.org/plosone/ article?id=10.1371/journal.pone.0029326. P.M. Catalano, Obesity, insulin resistance, and pregnancy outcome, Reproduction
Pregnancy Hypertension 15 (2019) 134–140
C.J. Petry et al.
[22]
[23]
[24]
[25]
[26]
[27]
[28]
140 (2010) 365–371 http://www.reproduction-online.org/content/140/3/365. long. C.N. Lumeng, A.R. Saltiel, Inflammatory links between obesity and metabolic disease, J. Clin. Invest. 121 (2011) 2111–2117 https://www.jci.org/articles/view/ 57132. H. Sell, C. Habich, J. Eckel, Adaptive immunity in obesity and insulin resistance, Nat. Rev. Endocrinol. 8 (2012) 709–716 https://www.nature.com/articles/nrendo. 2012.114. M.S. Zhou, A. Wang, H. Yu, Link between insulin resistance and hypertension: What is the evidence from evolutionary biology? Diabetol. Metab. Syndr. 6 (2014) 12 https://dmsjournal.biomedcentral.com/articles/10.1186/1758-5996-6-12. K.B. Clancy, L.D. Klein, A. Ziomkiewicz, I. Nenko, G. Jasienska, R.G. Bribiescas, Relationships between biomarkers of inflammation, ovarian steroids, and age at menarche in a rural Polish sample, Am. J. Hum. Biol. 25 (2013) 389–398 https:// onlinelibrary.wiley.com/doi/abs/10.1002/ajhb.22386. F. Juul, V.W. Chang, P. Brar, N. Parekh, Birth weight, early life weight gain and age at menarche: a systematic review of longitudinal studies, Obes. Rev. 18 (2017) 1272–1288 https://onlinelibrary.wiley.com/doi/abs/10.1111/obr.12587. L.G. Grunnet, S. Hansen, L. Hjort, C.M. Madsen, F.B. Kampmann, A.C.B. Thuesen, C. Granstrømi, M. Strøm, E. Maslova, R. Frikke-Schmidt, P. Damm, J.E. Chavarro, F.B. Hu, S.F. Olsen, A. Vaag, Adiposity, dysmetabolic traits, and earlier onset of female puberty in adolescent offspring of women with gestational diabetes mellitus: a clinical study within the Danish National Birth Cohort, Diabetes Care 40 (2017) 1746–1755 http://care.diabetesjournals.org/content/40/12/1746.long. I. Bralić, H. Tahirović, D. Matanić, O. Vrdoljak, S. Stojanović-Spehar, V. Kovacić,
[29]
[30]
[31]
[32]
[33]
[34]
140
S. Blazeković-Milaković, Association of early menarche age and overweight/obesity, J. Pediatr. Endocrinol. Metab. 25 (2012) 57–62 https://www.degruyter.com/ view/j/jpem.2012.25.issue-1-2/jpem-2011-0277/jpem-2011-0277.xml. K.S. Shim, Pubertal growth and epiphyseal fusion, Ann. Pediatr. Endocrinol. Metab. 20 (2015) 8–12 https://e-apem.org/journal/view.php?doi=10.6065/apem.2015. 20.1.8. R. Lakshman, N.G. Forouhi, S.J. Sharp, R. Luben, S.A. Bingham, K.T. Khaw, N.J. Wareham, K.K. Ong, Early age at menarche associated with cardiovascular disease and mortality, J. Clin. Endocrinol. Metab. 94 (2009) 4953–4960 https:// academic.oup.com/jcem/article/94/12/4953/2597000. K.K. Ong, D.B. Dunger, Perinatal growth failure: the road to obesity, insulin resistance and cardiovascular disease in adults, Best Pract. Res. Clin. Endocrinol. Metab. 16 (2002) 191–207 https://www.sciencedirect.com/science/article/pii/ S1521690X02901958?via%3Dihub. P. Prentice, C.L. Acerini, A. Eleftheriou, I.A. Hughes, K.K. Ong, D.B. Dunger, Cohort profile: the Cambridge Baby Growth Study (CBGS), Int. J. Epidemiol. 45 (2016) 35.a-g https://academic.oup.com/ije/article/45/1/35/236371. K.K. Ong, M.L. Ahmed, D.B. Dunger, Lessons from large population studies on timing and tempo of puberty (secular trends and relation to body size): the European trend, Mol. Cell. Endocrinol. 254–255 (2006) 8–12 https://www. sciencedirect.com/science/article/pii/S030372070600219X?via%3Dihub. R. Cooper, M. Blell, R. Hardy, S. Black, T.M. Pollard, M.E. Wadsworth, M.S. Pearce, D. Kuh, Validity of age at menarche self-reported in adulthood, J. Epidemiol. Community Health 60 (2006) 993–997 https://jech.bmj.com/content/60/11/993. long.